Published on Web 01/05/2010
Modulation of Electron Injection in CdSe-TiO2 System through Medium Alkalinity Vidhya Chakrapani, Kevin Tvrdy, and Prashant V. Kamat* Radiation Laboratory, Department of Chemistry and Biochemistry, and Department of Chemical and Biomolecular Engineering, UniVersity of Notre Dame, Notre Dame, Indiana 46556 Received November 13, 2009; E-mail:
[email protected] Photosensitization of TiO2 nanostructures with semiconducting quantum dots is useful in the development of all inorganic type solar cells.1,2 Previous studies have shown that the charge injection from excited CdSe nanocrystals into TiO2 can be tuned by controlling the CdSe particle size.3 Decreasing the particle diameter increases the driving force for the charge injection as the energy difference between the two conduction bands increases. For example, decreasing the particle diameter of CdSe from 7.5 to 2.8 nm resulted in a 3 orders of magnitude increase in the charge injection rate. An alternate way to modulate the energy difference between CdSe and TiO2 conduction bands is to tune the band edge of TiO2. Since it is difficult to obtain quantized TiO2 particles with varying size (particle diameter less than 1 nm), one needs to seek alternate means to tune the conduction band edge of TiO2.4,5 It has been established earlier that the pH-induced protonation of surface groups is useful to shift the band edges of TiO2 (a Nernstian shift of 59 mV/pH unit).6,7 The band edges shift to more negative (vs. NHE) potentials with increasing pH. Prior study on a dye sensitized TiO2 surface has shown that the oxidation potential of the dye itself changes with pH upon complexation with the surface; therefore it does not induce a major energy difference between the band edge of TiO2 and the excited state oxidation potential of the dye.8 In this regard, CdSe passivated with hydrophobic functional groups such as trioctyl phosphine oxide (TOPO) is useful as it renders the semiconductor surface insensitive to pH. Thus, any change in the solution pH only affects the energetic band position of TiO2. We have now capitalized on this ability to tune the band edge of TiO2 and established the importance of energy gap on the charge injection process between excited CdSe and TiO2. Here, we discuss modulation of the emission response of the CdSe-TiO2 system in the pH range 5-12. Nanostructured TiO2 and SiO2 films were prepared on fluorinedoped tin oxide (FTO) glass slides using respective colloidal nanoparticle suspensions. TOPO-capped CdSe quantum dots (3.5 nm in diameter) in toluene were prepared using a previously reported procedure.9 See Supporting Information (SI) for more details on the deposition of CdSe nanoparticles on TiO2 and SiO2 films. Fluorescence spectra were recorded using a 1 cm quartz cuvette that has provision to flow water of desired pH. All measurements were done by placing the films at 45° angle between the excitation source and the detector. Figure 1A shows the emission spectra of 3.5 nm CdSe nanocrystals deposited on TiO2 films that are in contact with solutions of three different pH values. While the TOPO-capped CdSe deposited on SiO2 substrate show little or no dependence (see Figure S2 in SI), we observe an increase in CdSe emission on TiO2 substrate as we increase the pH. As shown earlier, SiO2 is an inert substrate and does not directly influence the radiative recombination process in CdSe.10 On the other hand, the TiO2 substrate participates 1228
9
J. AM. CHEM. SOC. 2010, 132, 1228–1229
Figure 1. (A) Emission spectra and (B) emission response of CdSe on TiO2 film at three different pH values. Distilled water was flowed through the cell before changing pH. (C and D) Emission lifetime response of CdSe on TiO2 and SiO2 films at varying pH.
in a charge transfer interaction with excited CdSe. The fact that we observe an increased emission with increasing medium pH only in the case of TiO2 indicates that this effect is likely to arise from the altered rates of charge transfer. Interestingly, this change in the emission response is reversible if we allow enough time to equilibrate with the solution of different pH. Figure 1B shows the emission response monitored at 560 nm of CdSe-TiO2 film during the flow of solution of varying pH through the cell. The prompt response of the CdSe emission to the pH of the medium as well as its reversibility opens up the possibility of using CdSe-TiO2 systems for sensing medium pH. Unlike previous work, the present work does not employ dye or Förster resonance energy transfer (FRET mechanism) for sensing of pH.11 We further probed the influence of pH on the charge injection process by monitoring the emission decay of CdSe-TiO2 and CdSe-SiO2 films submerged in an aqueous solution of known pH (Figure 1C and D). The emission of CdSe-TiO2 film exhibits slower decay with increasing pH and, thus, confirms a longer radiative lifetime in alkaline pH. The emission decay exhibited a relatively smaller variance in the same pH range for CdSe-SiO2 films. We also further confirmed the reproducibility of the emission decay response to pH by comparing the emission-time profile at pH 5.5 before and after exposure to a solution of pH 12 (traces a and e in Figure 1C, respectively). Trace e was recorded after washing the TiO2 films with a copious amount of distilled water. The similarity of the emission decay lifetimes further confirms the reproducibility of the pH effect on the radiative recombination of CdSe deposited on TiO2 film. 10.1021/ja909663r 2010 American Chemical Society
COMMUNICATIONS
Figure 2. Dependence of electron injection rate constant, ket, on the pH for a TiO2-CdSe film (see SI for analysis).
The emission decay was analyzed by a three exponential fit. The kinetic parameters are summarized in the Supporting Information (Table S1). The average lifetime of CdSe-TiO2 film increased nearly 2-fold (from 3.17 to 6.20 ns) as we increased the pH from 5.5 to 12. Under similar experimental conditions, CdSe quantum dots attached to SiO2 exhibit little variation in the average emission lifetime (6.84-7.58 ns). The change in average lifetime agrees with the trend observed with the emission yield (Figure 1A). On SiO2 substrate, the CdSe quantum dots are nonresponsive to pH since radiative charge recombination is the only dominant process. However, on the TiO2 semiconductor surface the electron transfer to TiO2 competes with charge recombination and its dependence on pH causes the emission to vary. As shown earlier,10 we can use the average lifetime of the CdSe emission on SiO2 and TiO2 substrates to obtain an apparent electron transfer rate constant (ket). The ket values determined at different pH values are plotted in Figure 2. See Table S2 in the SI for details. A nearly 5-fold increase in the apparent electron transfer rate constant is seen upon decreasing the pH from 12 to 5.5. Figure 3 shows the energy levels of CdSe and TiO2 at two different pH values. The value of the conduction band edge of TiO2 at pH 12 and 5.5 is assumed to be -0.78 and -0.4 V vs NHE, respectively.7 For 3.5 nm diameter CdSe colloids, the conduction band is assumed to be -1.4 versus NHE.12 The band energies of TOPO-capped CdSe dots are independent of pH while the conduction band of TiO2 shifts to more negative potentials with increasing pH. As a result of this altered energetics of the CdSe-TiO2 system, we anticipate a decreased band energy difference to retard the rate of electron transfer from CdSe to TiO2. For pH 5.5 and 12 this energy difference (-∆G) decreases from ∼1.0 to ∼0.61 eV and a drop in the rate constant of 2 × 108 s-1 to 3 × 107 s-1 occurs respectively. The increased rate constants with decreasing pH further supports our earlier argument3 that the energy difference between the conduction bands of CdSe and TiO2 is important to attain an efficient charge injection process, and hence higher photoconversion efficiency. The modulation of the energy difference between conduction bands of CdSe and TiO2 was achieved through size quantization.3 In the present study we varied the energy difference by varying the band energy of TiO2 by means of medium pH while maintaining the same size of quantum dot. According to Marcus theory,13-15 for a nonadiabatic reaction in the activation limit, the rate of electron transfer depends on the electronic coupling between the donor and acceptor states, the density of states (DOS) per unit volume, and the driving force, which is the Gibbs free energy difference (-∆G) between the two states. Hence, as the driving force increases, the rate of charge transfer also increases until it reaches a maximum when the driving force is equal to the reorganizational energy. Because of the quasi
Figure 3. Scheme illustrating the electron injection processes from CdSe
band edge to TiO2 nanoparticle conduction band at two different pH conditions.
continuum of states in the TiO2 conduction band, the total electron transfer rate depends on the sum of all possible electronic transitions. Lian and co-workers have calculated the DOS dependent ket for Re-bipyridyl complexed to TiO2 and shown that electron transfer rate modulation per unit pH is greatest near the band edge and becomes nearly invariant for ∆G > 1 eV.16 Our experimental data are consistent with their observation. At lower pH, the greater availability of DOS is likely to make the electron transfer more efficient. Preliminary measurements on CdSe-TiO2 films at different pH values using femtosecond transient absorption spectroscopy show a trend similar to that for the emission decay results (see Figure S6). Efforts are currently underway to probe these processes at different excitation wavelengths and elucidate the role of DOS in the charge injection process. The ability to promote the charge injection from excited CdSe into TiO2 has important implications in the design of quantum dot sensitized solar cells. The results presented here demonstrate the modulation of the charge injection process by means of solution pH. The reversible emission response of the CdSe-TiO2 system to solution pH is also of interest in monitoring solution alkalinity. Acknowledgment. The research described herein was supported by the Office of Basic Energy Science of the Department of the Energy. V.C. also acknowledges the support of ND Energy Center for providing the research fellowship. This is contribution NDRL4837 from the Notre Dame Radiation Laboratory. Supporting Information Available: Experimental details, emission spectra of CdSe on silica, and emission lifetime analysis are presented. This material is available free of charge via the Internet at http:// pubs.acs.org. References (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16)
Nozik, A. J. Physica E 2002, 14, 115. Kamat, P. V. J. Phys. Chem. C 2008, 112, 18737. Robel, I.; Kuno, M.; Kamat, P. V. J. Am. Chem. Soc. 2007, 129, 4136. Satoh, N.; Nakashima, T.; Kamikura, K.; Yamamoto, K. Nat. Nanotechnol. 2008, 3, 106. Sant, P. A.; Kamat, P. V. Phys. Chem. Chem. Phys. 2002, 4, 198. Tomkiewicz, M. J. Electrochem. Soc. 1979, 126, 1505. Bolts, J. M.; Wrighton, M. S. J. Phys. Chem. 1976, 80, 2641. Zaban, A.; Ferrere, S.; Sprague, J.; Gregg, B. A. J. Phys. Chem. 1997, 101, 55. Peng, Z. A.; Peng, X. J. Am. Chem. Soc. 2001, 123, 183. Tvrdy, K.; Kamat, P. V. J. Phys. Chem. A 2009, 113, 3765. Snee, P. T.; Somers, R. C.; Nair, G.; Zimmer, J. P.; Bawendi, M. G.; Nocera, D. G. J. Am. Chem. Soc. 2006, 128, 13320. Wang, C. J.; Shim, M.; Guyot-Sionnest, P. Science 2001, 291, 2390. Marcus, R. A. J. Chem. Phys. 1965, 43, 679. Gao, Y. Q.; Yi Qui, G.; Georgievskii, Y.; Marcus, R. A. J. Chem. Phys. 2000, 112. Gao, Y. Q.; Marcus, R. A. J. Chem. Phys. 2000, 113, 6351. She, C.; Anderson, N. A.; Guo, J.; Liu, F.; Goh, W.-H.; Chen, D.-T.; Mohler, D. L.; Tian, Z.-Q.; Hupp, J. T.; Lian, T. J. Phys. Chem. B 2005, 109, 19345.
JA909663R J. AM. CHEM. SOC.
9
VOL. 132, NO. 4, 2010
1229
